Not only have there been many large shocks in this region during the past century (eight M≥6.5 events within 80 km, a rate of one a decade), but they often strike in pairs. M=6.5 and M=6.8 events stuck one day apart in 1953, and M=6.4 and M=6.5 events struck two months apart in 1948 on either side of the epicenter of the 2015 shock. Thus, the possibility that the 17 November shock could be part of a twin cannot be discounted.
The 17 November 2015 Mw=6.5 Nidri Greece earthquake struck in a highly active region in northwest Greece, at the junction of three active faults. Broadly, the ‘Nubia’ tectonic plate is impinging on Greece from the east at a rate of 6 millimeters per year, shoving the crust of the Ionian Sea beneath mainland Greece.
Large earthquakes often start or strike at the intersection of faults, as appears occurred here. The plate boundary is fractured into several active faults, including the Kephallonia fault to the south of the epicenter, which slips in a right-lateral sense at a rate of 5-20 mm/yr. (‘Right-lateral’ means that whatever side of the fault you are on, the other side slides to the right). For comparison, the San Andreas (California), and North Anatolia (Turkey) faults slip at about 25 mm/yr. Extending north of the epicenter is the right-lateral Lefkada fault, slipping at 4-8 mm/yr in a right-lateral sense. Intersecting the epicenter from the northwest is the Kerkya thrust fault, slipping 1-2 mm/yr. The mechanism of this quake suggests that the fault ruptured the Kephallonia-Lefkada system, with some thrust motion as well.
[Data from USGS, SHARE, and the ISC-GEM seismic catalog; map base from Google Earth. Interpretation by Ross S. Stein, Temblor, Inc. Nov 17, 2015]